KR20240029829A - Condition based maintenance prediction system using image big data for vehicle bogie sub-units in railway vehicles - Google Patents

Abstract

The present invention relates to a condition-based maintenance prediction system using video big data for the lower part of a railway vehicle bogie, and is a machine that photographs the exterior of the bogie, including wheels, axles, brake discs, and axle bearings, and provides exterior shooting data. vision department; A vibration measuring unit that measures vibration of the wheel and axle bearing and provides vibration measurement data; A thermal imaging unit that provides thermal image data by photographing the wheel, axle, and axle bearing; And by collecting the external shooting data, detecting abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and detecting abnormal vibrations in the wheels and axle bearings using the vibration measurement data. And, by including a collection and detection unit that detects abnormally high temperatures for the brake disc and axle bearing using the thermal image data, not only can defects in parts of the railway vehicle be effectively detected, but also deformation in the parts. , it is possible to predict and predict breakdown, dropout, and abnormal heat generation.

Description

Condition-based maintenance prediction system using image big data for railway vehicle bogie lower units {CONDITION BASED MAINTENANCE PREDICTION SYSTEM USING IMAGE BIG DATA FOR VEHICLE BOGIE SUB-UNITS IN RAILWAY VEHICLES}

The present invention collects external photographing data of a bogie including wheels, axles, brake discs, and axle bearings through a machine vision unit, detects abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and detects vibration. Abnormal vibration of the wheels and axle bearings is detected using vibration measurement data on the wheels and axle bearings measured through the measurement unit, and thermal image data on the wheels, axles, and axle bearings captured through the thermal imaging unit is used. By detecting abnormally high temperatures for brake discs and axle bearings, it is possible to effectively detect defects in parts of railway vehicles, as well as predict and predict deformation, failure, dislodgement, and abnormal heating of parts. This is about a condition-based maintenance prediction system using video big data for vehicle bogie lower units.

As is well known, railway vehicles are used as a major means of transportation not only because they can transport large quantities of cargo or passengers, but also because they are highly safe. Due to their transport scale, these railway vehicles use various techniques to reduce the possibility of accidents occurring. After detection, countermeasures must be established.

In addition, since railway vehicles run along rails, railway accidents can occur due to aging, defects, or damage to parts. This is mainly caused by coupling occurring in axles, wheels, etc., and the condition of the axles, wheels, etc. needs to be detected in advance.

Because such defects occur in various forms, they are difficult to identify through simple external inspection, and because multiple axles and wheels are combined in one railroad car, it is also difficult to conduct detection inspection simultaneously.

In relation to the detection and diagnosis of railway vehicle parts as described above, a technology was proposed to detect defects in railway vehicles by detecting the entry speed and vibration frequency of the railway vehicle. This can be done by detecting abnormal conditions in the vehicle and performing wireless communication with the management system.

However, defects in railway vehicle parts can occur in various forms, and operating conditions can change depending on the track or surrounding environment and defects, so it is necessary to detect parts defects in various ways, and technology for this is needed. Development is urgently needed.

1. Korean Patent No. 10-0388535 (registered on June 10, 2003)

The present invention collects external photographing data of a bogie including wheels, axles, brake discs, and axle bearings through a machine vision unit, detects abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and detects vibration. Abnormal vibration of the wheels and axle bearings is detected using vibration measurement data on the wheels and axle bearings measured through the measurement unit, and thermal image data on the wheels, axles, and axle bearings captured through the thermal imaging unit is used. By detecting abnormally high temperatures for brake discs and axle bearings, it is possible to effectively detect defects in parts of railway vehicles, as well as predict and predict deformation, failure, dislodgement, and abnormal heating of parts. We aim to provide a condition-based maintenance prediction system using video big data for vehicle bogie lower units.

The purposes of the embodiments of the present invention are not limited to the purposes mentioned above, and other purposes not mentioned will be clearly understood by those skilled in the art from the description below. .

According to an embodiment of the present invention, a machine vision unit for photographing the outer appearance of a bogie including wheels, axles, braking discs, and axle bearings to provide outer photographing data; A vibration measuring unit that measures vibration of the wheel and axle bearing and provides vibration measurement data; A thermal imaging unit that provides thermal image data by photographing the wheel, axle, and axle bearing; And by collecting the external shooting data, detecting abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and detecting abnormal vibrations in the wheels and axle bearings using the vibration measurement data. and a collection and detection unit that detects abnormally high temperatures for the brake disc and axle bearing using the thermal image data; a condition-based maintenance prediction system using image big data for the lower bogie device of a bogie railway vehicle, including a can be provided.

In addition, according to an embodiment of the present invention, the collection and detection unit counts each abnormal condition for the wheel, axle, brake disc, and axle bearing, and when the counting number is relatively larger than the preset reference value, the detection unit counts the abnormal condition corresponding to the replacement time. A condition-based maintenance prediction system using video big data for the undercarriage device of a railroad car that provides warning alarms can be provided.

In addition, according to an embodiment of the present invention, the machine vision unit includes a 2D camera, a 3D camera, lighting, an illumination sensor, and a speed sensor, and detects the speed and illumination of the cart to use the 2D camera, the 3D camera, and the lighting. A condition-based maintenance prediction system using video big data for the controlled railway vehicle bogie lower unit can be provided.

In addition, according to an embodiment of the present invention, the vibration measuring unit includes a vibration accelerometer, a signal converter, and a data link, and measures the vibration of the bogie and transmits the vibration measurement data to the collection and detection unit through the signal converter and the data link. A condition-based maintenance prediction system can be provided using video big data for the underbody device of a railway vehicle bogie that provides.

In addition, according to an embodiment of the present invention, the thermal image capturing unit includes a thermal imaging camera, an optical converter, and a camera link, and captures a thermal image of the truck and transmits the thermal image to the collection and detection unit through the optical converter and the camera link. A condition-based maintenance prediction system using image big data for the underbody device of a railway vehicle bogie that provides the thermal image data may be provided.

In addition, according to an embodiment of the present invention, the collection and detection unit compares 3D image data and reference image data among the external shooting data through the artificial intelligence algorithm to determine whether there is an abnormality in the wheel, axle, brake disc, and axle bearing. A condition-based maintenance prediction system using image big data for each railway vehicle bogie lower device can be provided.

In addition, according to an embodiment of the present invention, the collection and detection unit performs condition-based maintenance using image big data for a railway vehicle bogie lower device to realign images according to speed before comparing the 3D image data and reference image data. A prediction system may be provided.

In addition, according to an embodiment of the present invention, the collection and detection unit performs a difference operation on the 3D image data and the reference image data to detect parts that are different from the reference image data. A condition-based maintenance prediction system using data can be provided.

In addition, according to an embodiment of the present invention, abnormal heat failure data of the brake disc and axle bearing detected by the condition-based maintenance prediction system and 2D images of failure and detachment of the wheel, axle, brake disc, and axle bearing A railway vehicle bogie lower unit that builds big data based on data and 3D image data and applies the artificial intelligence algorithm to predict deformation, failure, dislodgement, and abnormal heat for the wheels, axles, brake discs, and axle bearings. A condition-based maintenance prediction system using video big data can be provided.

The present invention collects external photographing data of a bogie including wheels, axles, brake discs, and axle bearings through a machine vision unit, detects abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and detects vibration. Abnormal vibration of the wheels and axle bearings is detected using vibration measurement data on the wheels and axle bearings measured through the measurement unit, and thermal image data on the wheels, axles, and axle bearings captured through the thermal imaging unit is used. By detecting abnormally high temperatures for brake discs and axle bearings, defects in parts of railway vehicles can be effectively detected, as well as deformation, failure, detachment, and abnormal heating of parts can be predicted and predicted.

Figure 1 is a block diagram of a condition-based maintenance prediction system using image big data for a railway vehicle bogie lower device according to an embodiment of the present invention;
Figure 2 is a diagram conceptually illustrating a condition-based maintenance prediction system using image big data for a railway vehicle bogie lower device according to an embodiment of the present invention;
Figures 3 to 13 are diagrams for explaining detection of abnormalities in wheels, axles, brake discs, and axle bearings according to an embodiment of the present invention;
Figures 14 to 16 are diagrams to explain detecting abnormalities in wheels, axles, brake discs, and axle bearings according to an embodiment of the present invention.

Advantages and features of the embodiments of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to be understood by those skilled in the art. It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

In describing embodiments of the present invention, if a detailed description of a known function or configuration is judged to unnecessarily obscure the gist of the present invention, the detailed description will be omitted. The terms described below are terms defined in consideration of functions in the embodiments of the present invention, and may vary depending on the intention or custom of the user or operator. Therefore, the definition should be made based on the contents throughout this specification.

Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

Figure 1 is a block diagram of a condition-based maintenance prediction system using image big data for a railway vehicle bogie lower device according to an embodiment of the present invention, and Figure 2 is a block diagram of a condition-based maintenance prediction system for a railroad vehicle bogie lower device according to an embodiment of the present invention. It is a drawing conceptually illustrating a condition-based maintenance prediction system using image big data, and Figures 3 to 13 are diagrams for detecting abnormalities in wheels, axles, brake discs, and axle bearings according to an embodiment of the present invention. Figures 14 to 16 are diagrams for explaining the detection of abnormalities in wheels, axles, brake discs, and axle bearings according to an embodiment of the present invention.

Referring to FIGS. 1 to 16, the condition-based maintenance prediction system using image big data for the underbody device of a railway vehicle bogie according to an embodiment of the present invention includes a machine vision unit 110, a vibration measurement unit 120, and a heat It may include an image capturing unit 130, a collection and detection unit 140, etc.

The machine vision unit 110 provides external shooting data by photographing the external appearance of the bogie, including wheels, axles, brake discs, and axle bearings, and includes a 2D camera, a 3D camera, lighting, an illumination sensor, and a speed sensor. By detecting the speed and illumination of the cart, you can control 2D cameras, 3D cameras, and lighting.

For example, as shown in FIG. 2, the machine vision unit 110 uses at least one 2D camera, at least one 3D camera, and at least one light centered on the rail on the path along which the railroad car moves along the rail. And, at least one illuminance sensor and at least one speed sensor may be disposed.

Here, the railway vehicle can pass through the detection area at a maximum speed of approximately 60 km/h, using at least one 2D camera and at least one 3D camera, as shown in FIGS. 3 and 4. By photographing the wheels, axles, brake discs, and axle bearings of the bogie, the exterior photographing data can be provided to the collection and detection unit 140.

At this time, in order to improve image clarity according to environmental factors such as weather and locational factors of the bogie located at the bottom of the railroad car, the output intensity of the light is adjusted according to the illuminance measured through the illuminance sensor to configure the bogie. The external appearance of wheels, axles, brake discs, and axle bearings can be effectively photographed.

Meanwhile, the machine vision unit 110 can measure the speed at which a railway vehicle enters through a speed sensor and provide the speed data to the collection and detection unit 140 to realign images according to speed.

The vibration measurement unit 120 provides vibration measurement data by measuring the vibration of the wheels and axle bearings, and includes a vibration accelerometer, a signal converter, and a data link. It measures the vibration of the bogie and provides vibration measurement data through the signal converter and data link. Vibration measurement data can be provided to the collection and detection unit 140.

For example, the vibration measuring unit 120 may have at least one vibration accelerometer disposed along the rail on the path in which the railroad car moves along the rail, as shown in FIGS. 2 and 3, and the acceleration measured through this Information can be provided to the collection and detection unit 140 as vibration measurement data corresponding to the wheel and axle bearing through a signal converter and data link.

This vibration measurement unit 120 may additionally place at least one acoustic sensor (for example, a sound cam, etc.), and the acoustic data measured therefrom is included in the vibration measurement data and provided to the collection and detection unit 140. can do.

The thermal imaging unit 130 provides thermal image data by photographing the wheels, axles, and axle bearings, and includes a thermal imaging camera, an optical converter, and a camera link. It captures thermal images of the bogie and provides thermal image data to the optical converter and camera. Thermal image data can be provided to the collection and detection unit 140 through a link.

For example, as shown in FIGS. 2 and 3, the thermal imaging unit 130 includes at least one thermal imaging camera with an imaging direction pointing upward from the center of the rail on the path along which the railroad car moves along the rail. It can be arranged, and the thermal image captured therefrom can be provided to the collection and detection unit 140 as thermal image data for the wheel, axle, and axle bearing through an optical converter and camera link.

Meanwhile, as shown in FIG. 2, in addition to each sensor of the machine vision unit 110, vibration measurement unit 120, and thermal imaging unit 130 as described above, there is at least one sensor capable of recognizing the car number of the railway vehicle. A car sign recognition sensor and at least one detection sensor that detects the entry of a railway vehicle may be additionally provided.

The collection and detection unit 140 collects external shooting data, detects abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and uses vibration measurement data to determine whether there are abnormal vibrations in the wheels and axle bearings. Detects abnormally high temperatures for the brake disc and axle bearing using thermal image data. Each abnormal condition for the wheel, axle, brake disc, and axle bearing is counted, and the counting number is relative to the preset standard value. If the value is large, a warning alarm corresponding to the replacement period can be provided.

Here, the artificial intelligence algorithm may include, for example, a CNN (convolution neural network) algorithm. The CNN algorithm is a useful algorithm for finding images, and learns images directly from data and classifies images using patterns. It has the advantage of being able to effectively process images by learning while maintaining the spatial information of the image.

For example, as shown in FIG. 5, the collection and detection unit 140 performs artificial intelligence learning about normal and abnormal states using a plurality of images of the wheels, axles, brake discs, and axle bearings that make up the bogie. After performing the operation, the appearance shooting data provided from the machine vision unit 110 through the learned artificial intelligence algorithm is used to check whether there are any abnormalities in the appearance of the wheels, axles, brake discs, and axle bearings of the bogie (i.e., damage, cracks, etc.) dents, etc.) can be detected individually.

In addition, when speed data is provided from the machine vision unit 110, the collection and detection unit 140 provides a shooting control signal to the machine vision unit 110 to determine the shooting speed of the 2D camera and the 3D camera in response to the speed data (i.e. , shutter speed) can be controlled, and in addition, a light emission control signal that controls the emission of lighting in conjunction with the shutter speed can be provided to the machine vision unit 110.

Meanwhile, as shown in FIGS. 6 and 7, the collection and detection unit 140 uses thermal image data on the axle and axle bearing from the thermal imaging unit 130 to inspect the axle and the axle bearing for abnormalities. This can be done by receiving thermal image data for the axle and axle bearing, detecting the temperature of the axle and axle bearing, and It is possible to determine the normal or abnormal state of the axle bearing (i.e., recognize abnormal state patterns and predict the state of the axle and axle bearing, etc.).

And, as shown in FIGS. 8 to 10, the collection and detection unit 140 collects external imaging data for the braking disk from the machine vision unit 110 and thermal image data for the braking disk from the thermal imaging unit 130. The presence or absence of abnormalities in the braking disc can be checked using the collection and detection unit 140, which performs artificial intelligence learning on normal and abnormal states using a plurality of images of the braking disc, and then performs artificial intelligence learning on the learned artificial intelligence. Through an algorithm, the presence or absence of abnormalities in the external appearance of the braking disc can be detected using the external photographing data provided from the machine vision unit 110.

In addition, the collection and detection unit 140 receives thermal image data for the braking disc, detects the temperature of the braking disc, and detects the temperature according to the temperature analysis range (e.g., -40 to 55 ℃, resolution ±1 ℃). It is possible to determine the normal state or abnormal state of the braking disk (i.e., recognize the abnormal state pattern and predict the state of the braking disk, etc.).

And, as shown in FIGS. 11 to 13, the collection and detection unit 1140 collects the vibration measurement data of the wheel provided from the vibration measurement unit 120 and the thermal image data of the wheel provided from the thermal imaging unit 130. The presence or absence of abnormalities in the wheel can be checked by collecting the vibration frequency from the vibration measurement data of the wheel provided from the vibration measurement unit 120 and analyzing the discrete Fourier transform (for example, analysis range 0-50 G, The resolution is 25N).

In addition, the collection and detection unit 1140 receives thermal image data for the wheel, detects the temperature of the wheel, and detects the temperature of the wheel according to the temperature analysis range (e.g., -40 to 200 ℃, resolution ±1 ℃). It is possible to determine a normal state or an abnormal state (i.e., recognize an abnormal state pattern for a vehicle, predict the state, etc.).

The collection and detection unit 140 as described above can predict the replacement time of the corresponding part if the counting number for counting the presence or absence of an abnormality in any one selected among the wheel, axle, brake disc, and axle bearing is relatively larger than the preset reference value. In addition, a warning alarm corresponding to replacement time can be provided.

For example, if the standard value is set to correspond to immediate replacement, a warning alarm for immediate replacement can be provided if the counting number is relatively larger than the preset standard value, and if the standard value is set to correspond to replacement after 1 month, If the counting number is relatively larger than the preset standard value, a warning alarm can be provided indicating that it must be replaced within 1 month.

This checking of the counting number can be performed for each part or for all parts, and of course, it can be performed by classifying parts that exhibit similar characteristics.

Meanwhile, the detection accuracy of the bogie abnormality detection device according to the embodiment of the present invention is as shown in Table 1 below.

That is, in the case of vision inspection, the bogie abnormal condition detection device according to an embodiment of the present invention can detect and determine deformation and falloff by recognizing major parts on the bottom and side of the bogie, and recognizes the pattern of the braking disc. Detection and judgment on state prediction can be performed.

In addition, in the case of thermal imaging inspection, it is possible to detect and determine whether there is an abnormality by measuring and distinguishing the temperature in the temperature range of ??40 to 600 ℃ (resolution ±1 ℃) for the shaft bearing, and ??40 for the brake disc. It is possible to detect and determine whether there is an abnormality by measuring and distinguishing the temperature in the temperature range of 55 ℃ to 55 ℃ (resolution ±1 ℃).

In addition, in the case of vibration inspection, the presence or absence of abnormalities can be detected and judged by measuring and distinguishing the vibration frequency in the measurement range of 0-50 g (25N) for the axle bearing, and the presence or absence of abnormalities can be detected and judged in the measurement range of 0-50 g (25N) for the wheel. ), you can detect and determine whether there is an abnormality by measuring and distinguishing the vibration frequency.

Meanwhile, the collection and detection unit 140 as described above can detect abnormalities in the wheels, axles, brake discs, and axle bearings by comparing 3D image data and reference image data among the external shooting data through an artificial intelligence algorithm. , Before comparing 3D image data and reference image data, images can be rearranged according to speed, and differences from the reference image data can be detected by performing a difference operation on the 3D image data and reference image data.

For example, as shown in FIG. 14, the collection and detection unit 140 normalizes and pre-processes the 3D image data provided from the machine vision unit 110 and the previously stored reference image data. After defining the starting point (head) of the vehicle and extracting the histogram, which is the data frequency distribution table of the image corresponding to the starting point, the part where the histogram matches can be detected in the 3D image data through a head finder. In addition, when a starting point is detected in 3D image data, the reference image data can be divided and rearranged.

That is, as shown in the left drawing of FIG. 15, the starting point (i.e., upper reference image data) found through the head finder and the lower (3D image data) starting point (vertical line portion) can be rearranged to match.

In addition, the appearance shooting data of the bogie due to the speed of the railway vehicle causes the image to stretch or shrink. To solve this, the changing speed data is provided from the machine vision unit 110 and the image is rearranged. Speed calibration can be performed in this way. That is, as shown in the right drawing of FIG. 15, assuming a situation of deceleration, the stretched image can be corrected to match the original speed.

Next, as shown in FIG. 16, the collection and detection unit 140 can perform difference calculation on the pre-processed 3D image data and the reference image data, and compare the 3D image data with the reference image data. It is possible to display areas where the pixel values of the data differ, and by differentiating the image to express contour lines, it is possible to detect areas where changes in pixel values clearly occur (contours).

Accordingly, the collection and detection unit 140 can effectively detect and detect abnormal condition bogie parts by correcting the external shooting image according to the speed of the railway vehicle and then comparing it with the reference image data.

As described above, in the embodiment of the present invention, abnormal heat failure data of the brake disc and axle bearing detected by the condition-based maintenance prediction system, 2D image data for failure and separation of the wheel, axle, brake disc, and axle bearing, and By building big data based on 3D image data and applying artificial intelligence algorithms, it is possible to predict and predict deformation, failure, dislodgement, and abnormal heat for wheels, axles, brake discs, and axle bearings.

Therefore, according to an embodiment of the present invention, external photographing data of the bogie including the wheels, axles, braking discs, and axle bearings are collected through the machine vision unit, and the wheels, axles, brake discs, and axle bearings are captured through an artificial intelligence algorithm. Detects abnormalities, detects abnormal vibrations in wheels and axle bearings using vibration measurement data on wheels and axle bearings measured through the vibration measurement unit, and detects abnormal vibrations in the wheels, axles, and axle bearings captured through the thermal imaging unit. By using thermal imaging data to detect abnormally high temperatures for brake discs and axle bearings, defects in parts of railway vehicles can be effectively detected, as well as deformation, failure, fall-off, and abnormal heating of parts. Can predict and foresee.

In the above description, various embodiments of the present invention have been presented and explained, but the present invention is not necessarily limited thereto, and those skilled in the art will understand various embodiments without departing from the technical spirit of the present invention. It will be easy to see that branch substitutions, transformations, and changes are possible.

110: Machine vision unit 120: Vibration measurement unit
130: thermal imaging unit 140: collection and detection unit

Claims (9)

A machine vision unit that photographs the exterior of the bogie, including wheels, axles, brake discs, and axle bearings, and provides exterior imaging data;
A vibration measuring unit that measures vibration of the wheel and axle bearing and provides vibration measurement data;
A thermal imaging unit that provides thermal image data by photographing the wheel, axle, and axle bearing; and
Collect the external shooting data to detect abnormalities in the wheels, axles, brake discs, and axle bearings through an artificial intelligence algorithm, and use the vibration measurement data to detect abnormal vibrations in the wheels and axle bearings. , a collection and detection unit that detects abnormally high temperatures for the braking disc and axle bearing using the thermal image data;
A condition-based maintenance prediction system using video big data for the underbody device of a railway vehicle bogie, including.
In claim 1,
The collection and detection unit,
Counts each abnormal condition for the wheel, axle, brake disc, and axle bearing, and provides a warning alarm corresponding to replacement time if the counting number is relatively larger than the preset standard value.
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
In claim 2,
The machine vision department,
It includes a 2D camera, a 3D camera, lighting, an illumination sensor, and a speed sensor, and detects the speed and illumination of the cart to control the 2D camera, 3D camera, and lighting.
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
In claim 3,
The vibration measuring unit,
It includes a vibration accelerometer, a signal converter, and a data link, and measures the vibration of the bogie and provides the vibration measurement data to the collection and detection unit through the signal converter and data link.
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
In claim 4,
The thermal imaging unit,
It includes a thermal imaging camera, an optical converter, and a camera link, and takes thermal images of the cart and provides the thermal image data to the collection and detection unit through the optical converter and camera link.
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
In claim 5,
The collection and detection unit,
Comparing 3D image data and reference image data among the appearance shooting data through the artificial intelligence algorithm to detect abnormalities in the wheels, axles, brake discs, and axle bearings, respectively.
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
In claim 6,
The collection and detection unit,
Reordering images according to speed before comparing the 3D image data and reference image data
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
In claim 7,
The collection and detection unit,
Performing difference calculation on the 3D image data and reference image data to detect parts that are different from the reference image data
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.
The method according to any one of claims 1 to 8,
Big data based on abnormal heat generation failure data of the brake disc and axle bearing detected by the condition-based maintenance prediction system and 2D image data and 3D image data about failure and separation of the wheel, axle, brake disc, and axle bearing. Build and apply the artificial intelligence algorithm to predict deformation, failure, dropout, and abnormal heat for the wheels, axles, brake discs, and axle bearings.
Condition-based maintenance prediction system using video big data for railway vehicle bogie lower units.